Who should mark DSCP values in a LAN deployment?

• A. The core router
• B. The gateway at the data center
• C. The endpoint itself
• D. The closest trusted source, like endpoints or access switch

Answer: (D)

DSCP marking should be done as close to the source as possible on a device you trust to classify the traffic. In a LAN, that means at the edge—on the endpoint itself or on the access switch that directly connects to the endpoint. By marking at this trusted boundary, the DSCP value is preserved as traffic moves through the rest of the network, and the core devices can simply honor those marks rather than re-marking or altering them. This avoids inconsistent or conflicting classifications and ensures consistent QoS treatment across the LAN. Marking at the core or only at the data center gateway can lead to mis-marking or loss of the intended service class, whereas edge-based marking provides a reliable and scalable approach.